Change search
Refine search result
1234567 1 - 50 of 544
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf
Rows per page
  • 5
  • 10
  • 20
  • 50
  • 100
  • 250
Sort
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
  • Standard (Relevance)
  • Author A-Ö
  • Author Ö-A
  • Title A-Ö
  • Title Ö-A
  • Publication type A-Ö
  • Publication type Ö-A
  • Issued (Oldest first)
  • Issued (Newest first)
  • Created (Oldest first)
  • Created (Newest first)
  • Last updated (Oldest first)
  • Last updated (Newest first)
  • Disputation date (earliest first)
  • Disputation date (latest first)
Select
The maximal number of hits you can export is 250. When you want to export more records please use the Create feeds function.
  • 1.
    Aayesha, Aayesha
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Nouri, Jalal
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Afzaal, Muhammad
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Wu, Yongchao
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Li, Xiu
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Weegar, Rebecka
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    An Ensemble Approach for Question-Level Knowledge Tracing2021In: Artificial Intelligence in Education: 22nd International Conference, AIED 2021, Utrecht, The Netherlands, June 14–18, 2021, Proceedings, Part II / [ed] Ido Roll; Danielle McNamara; Sergey Sosnovsky; Rose Luckin; Vania Dimitrova, Cham: Springer , 2021, p. 433-437Conference paper (Refereed)
    Abstract [en]

    Knowledge tracing—where a machine models the students’ knowledge as they interact with coursework—is a well-established area in the field of Artificial Intelligence in Education. In this paper, an ensemble approach is proposed that addresses existing limitations in question-centric knowledge tracing and achieves the goal of predicting future question correctness. The proposed approach consists of two models; one is Light Gradient Boosting Machine (LightGBM) built by incorporating all relevant key features engineered from the data. The second model is a Multiheaded-Self-Attention Knowledge Tracing model (MSAKT) that extracts historical student knowledge of future question by calculating their contextual similarity with previously attempted questions. The proposed model’s effectiveness is evaluated by conducting experiments on a big Kaggle dataset achieving an Area Under ROC Curve (AUC) score of 0.84 with 84% accuracy using 10fold cross-validation.

  • 2. Abd-Alrhman, A.M.
    et al.
    Ekenberg, Love
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    A Multi-layered Data Preparation Model for Health Information in Sudan2020In: The International Journal on Advances in ICT for Emerging Regions, ISSN 1800-4156, Vol. 13, no 3, p. 1-14Article in journal (Refereed)
  • 3.
    Adetona, Temitayo Eniola
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Information Security Management and Organisational Agility2023Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    An organization's ability to succeed depends on the Confidentiality, Integrity, and Availability of its information. This implies that the organization's information and assets must be secured and protected. However, the regular occurrence of threats, risks, and intrusions could serve as a barrier to the security of this information. This has made the management of Information security a necessity. Organizations are then trying to be more agile by looking for ways to identify and embrace opportunities swiftly and confront these risks more quickly. Very little research has examined the relationships between Organizational Agility and Information Security. Hence, this study aims to investigate the management of Information Security in organizations while maintaining agility and highlighting the challenges encountered, and also addresses the research question: How do organizations manage information security while maintaining organizational agility?

    The research strategy used is the Case Study, and the data collection methods used are semi-structured interviews and documents. The interview was conducted in a financial institution in Nigeria with seven security specialists, and documents were obtained from the company's website to help gain insights into the services and products offered. Thematic analysis was the data analysis method chosen. The findings revealed eighteen measures in which Information Security can be managed while maintaining Organizational Agility. Part of the identified measures are similar to those identified in previous research, while new measures are also discovered. Furthermore, these identified measures will be useful for other organizations, particularly financial institutions, to emulate in managing their Information Security and being agile while at it.

    Download full text (pdf)
    FULLTEXT01
  • 4. Afshari, Bahareh
    et al.
    Enqvist, Sebastian
    Stockholm University, Faculty of Humanities, Department of Philosophy.
    Leigh, Graham e.
    Marti, Johannes
    Venema, Yde
    PROOF SYSTEMS FOR TWO-WAY MODAL MU-CALCULUS2023In: Journal of Symbolic Logic (JSL), ISSN 0022-4812, E-ISSN 1943-5886Article in journal (Refereed)
    Abstract [en]

    We present sound and complete sequent calculi for the modal mu-calculus with converse modalities, aka two-way modal mu-calculus. Notably, we introduce a cyclic proof system wherein proofs can be represented as finite trees with back-edges, i.e., finite graphs. The sequent calculi incorporate ordinal annotations and structural rules for managing them. Soundness is proved with relative ease as is the case for the modal mu-calculus with explicit ordinals. The main ingredients in the proof of completeness are isolating a class of non-wellfounded proofs with sequents of bounded size, called slim proofs, and a counter-model construction that shows slimness suffices to capture all validities. Slim proofs are further transformed into cyclic proofs by means of re-assigning ordinal annotations.

  • 5.
    Afzaal, Muhammad
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Nouri, Jalal
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Aayesha, Aayesha
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Papapetrou, Panagiotis
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Fors, Uno
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Wu, Yongchao
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Li, Xiu
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Weegar, Rebecka
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Automatic and Intelligent Recommendations to Support Students’ Self-Regulation2021In: International Conference on Advanced Learning Technologies (ICALT),, 2021, p. 336-338Conference paper (Refereed)
    Abstract [en]

    In this paper, we propose a counterfactual explanations-based approach to provide an automatic and intelligent recommendation that supports student's self-regulation of learning in a data-driven manner, aiming to improve their performance in courses. Existing work under the fields of learning analytics and AI in education predict students' performance and use the prediction outcome as feedback without explaining the reasons behind the prediction. Our proposed approach developed an algorithm that explains the root causes behind student's performance decline and generates data-driven recommendations for action. The effectiveness of the proposed predictive model that constitutes the intelligent recommendations is evaluated, with results demonstrating high accuracy.

  • 6.
    Afzaal, Muhammad
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Nouri, Jalal
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Aayesha, Aayesha
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Papapetrou, Panagiotis
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Fors, Uno
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Wu, Yongchao
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Li, Xiu
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Weegar, Rebecka
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Generation of Automatic Data-Driven Feedback to Students Using Explainable Machine Learning2021In: Artificial Intelligence in Education: 22nd International Conference, AIED 2021, Utrecht, The Netherlands, June 14–18, 2021, Proceedings, Part II / [ed] Ido Roll; Danielle McNamara; Sergey Sosnovsky; Rose Luckin; Vania Dimitrova, Springer , 2021, p. 37-42Conference paper (Refereed)
    Abstract [en]

    This paper proposes a novel approach that employs learning analytics techniques combined with explainable machine learning to provide automatic and intelligent actionable feedback that supports students self-regulation of learning in a data-driven manner. Prior studies within the field of learning analytics predict students’ performance and use the prediction status as feedback without explaining the reasons behind the prediction. Our proposed method, which has been developed based on LMS data from a university course, extends this approach by explaining the root causes of the predictions and automatically provides data-driven recommendations for action. The underlying predictive model effectiveness of the proposed approach is evaluated, with the results demonstrating 90 per cent accuracy.

  • 7.
    Afzaal, Muhammad
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Nouri, Jalal
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Zia, Aayesha
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Papapetrou, Panagiotis
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Fors, Uno
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Wu, Yongchao
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Li, Xiu
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Weegar, Rebecka
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Explainable AI for Data-Driven Feedback and Intelligent Action Recommendations to Support Students Self-Regulation2021In: Frontiers in Artificial Intelligence, E-ISSN 2624-8212, Vol. 4, article id 723447Article in journal (Refereed)
    Abstract [en]

    Formative feedback has long been recognised as an effective tool for student learning, and researchers have investigated the subject for decades. However, the actual implementation of formative feedback practices is associated with significant challenges because it is highly time-consuming for teachers to analyse students’ behaviours and to formulate and deliver effective feedback and action recommendations to support students’ regulation of learning. This paper proposes a novel approach that employs learning analytics techniques combined with explainable machine learning to provide automatic and intelligent feedback and action recommendations that support student’s self-regulation in a data-driven manner, aiming to improve their performance in courses. Prior studies within the field of learning analytics have predicted students’ performance and have used the prediction status as feedback without explaining the reasons behind the prediction. Our proposed method, which has been developed based on LMS data from a university course, extends this approach by explaining the root causes of the predictions and by automatically providing data-driven intelligent recommendations for action. Based on the proposed explainable machine learning-based approach, a dashboard that provides data-driven feedback and intelligent course action recommendations to students is developed, tested and evaluated. Based on such an evaluation, we identify and discuss the utility and limitations of the developed dashboard. According to the findings of the conducted evaluation, the dashboard improved students’ learning outcomes, assisted them in self-regulation and had a positive effect on their motivation.

    Download full text (pdf)
    fulltext
  • 8. Ahlkrona, Josefin
    et al.
    Lötstedt, Per
    Kirchner, Nina
    Stockholm University, Faculty of Science, Department of Physical Geography.
    Zwinger, Thomas
    Dynamically coupling the non-linear Stokes equations with the shallow ice approximation in glaciology: Description and first applications of the ISCAL method2016In: Journal of Computational Physics, ISSN 0021-9991, E-ISSN 1090-2716, Vol. 308, p. 1-19Article in journal (Refereed)
    Abstract [en]

    We propose and implement a new method, called the Ice Sheet Coupled Approximation Levels (ISCAL) method, for simulation of ice sheet flow in large domains during long time-intervals. The method couples the full Stokes (FS) equations with the Shallow Ice Approximation (SIA). The part of the domain where SIA is applied is determined automatically and dynamically based on estimates of the modeling error. For a three dimensional model problem, ISCAL computes the solution substantially faster with a low reduction in accuracy compared to a monolithic FS. Furthermore, ISCAL is shown to be able to detect rapid dynamic changes in the flow. Three different error estimations are applied and compared. Finally, ISCAL is applied to the Greenland Ice Sheet on a quasi-uniform grid, proving ISCAL to be a potential valuable tool for the ice sheet modeling community.

  • 9.
    Ahlqvist, Oskar
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Detecting Fraud in Affiliate Marketing: Comparative Analysis of Supervised Machine Learning Algorithms2023Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
    Abstract [en]

    Affiliate marketing has become a rapidly growing part of the digital marketing sector. However, fraud in affiliate marketing raises a serious threat to the trust and financial stability of the involved parties. This thesis investigates the performance of three supervised machine learning algorithms - random forest, logistic regression, and support vector machine in detecting fraud in affiliate marketing. The objective is to answer the following main research question by answering two sub-questions:

    How much can Random Forest, Logistic Regression, and Support Vector Machine contribute to the detection of fraud in affiliate marketing?

    1. How can the models be compared in an experiment?

    2. How can they be optimized and applied within an affiliate marketing framework?

    To answer these questions, a dataset of transaction logs is analyzed in collaboration with an affiliate network company. The machine learning experiment employs k-fold crossvalidation and the Area Under the ROC Curve (AUC-ROC) performance metric to evaluate the effectiveness of the classifiers in distinguishing fraudulent from non-fraudulent transactions.

    The results indicate that the random forest classifier performs best out of the models, achieving the highest mean AUC of 0.7172. Furthermore, using feature importance analysis demonstrates that each feature category had different impact on the performance of the models. It was discovered that the models computes different feature importance meaning that some features displayed greater influence on specific models. By fine-tuning and optimizing the hyperparameters for each model, it is possible to enhance their performance.

    Despite certain limitations, such as time constraints, data availability, and security restrictions, this study highlights the potential of supervised machine learning algorithms. Particularly random forest showed to how it could be used to improve fraud detection capabilities in affiliate marketing.The insights contribute to closing the knowledge gap in comparing the effectiveness of various classification methods and practical applications for fraud detection.

    Download full text (pdf)
    FULLTEXT01
  • 10. Ahrens, Benedikt
    et al.
    Huber, Simon
    Mörtberg, Anders
    Stockholm University, Faculty of Science, Department of Mathematics.
    Preface to the MSCS Issue 31.1 (2021) Homotopy Type Theory and Univalent Foundations2021In: Mathematical Structures in Computer Science, ISSN 0960-1295, E-ISSN 1469-8072, Vol. 31, no 1, p. 1-2Article in journal (Other academic)
  • 11. Ahrens, Benedikt
    et al.
    Lumsdaine, Peter Lefanu
    Stockholm University, Faculty of Science, Department of Mathematics.
    Displayed Categories2019In: Logical Methods in Computer Science, ISSN 1860-5974, E-ISSN 1860-5974, Vol. 15, no 1, article id 20Article in journal (Refereed)
    Abstract [en]

    We introduce and develop the notion of displayed categories. A displayed category over a category C is equivalent to 'a category D and functor F : D -> C', but instead of having a single collection of 'objects of D' with a map to the objects of C, the objects are given as a family indexed by objects of C, and similarly for the morphisms. This encapsulates a common way of building categories in practice, by starting with an existing category and adding extra data/properties to the objects and morphisms. The interest of this seemingly trivial reformulation is that various properties of functors are more naturally defined as properties of the corresponding displayed categories. Grothendieck fibrations, for example, when defined as certain functors, use equality on objects in their definition. When defined instead as certain displayed categories, no reference to equality on objects is required. Moreover, almost all examples of fibrations in nature are, in fact, categories whose standard construction can be seen as going via displayed categories. We therefore propose displayed categories as a basis for the development of fibrations in the type-theoretic setting, and similarly for various other notions whose classical definitions involve equality on objects. Besides giving a conceptual clarification of such issues, displayed categories also provide a powerful tool in computer formalisation, unifying and abstracting common constructions and proof techniques of category theory, and enabling modular reasoning about categories of multi-component structures. As such, most of the material of this article has been formalised in Coq over the UniMath library, with the aim of providing a practical library for use in further developments.

  • 12. Ahrens, Benedikt
    et al.
    Matthes, Ralph
    Mörtberg, Anders
    Stockholm University, Faculty of Science, Department of Mathematics.
    Implementing a category-theoretic framework for typed abstract syntax2022In: CPP '22: Proceedings of the 11th ACM SIGPLAN International Conference on Certified Programs and Proofs / [ed] Andrei Popescu; Steve Zdancewic, New York: Association for Computing Machinery (ACM), 2022, p. 307-323Conference paper (Refereed)
    Abstract [en]

    In previous work ("From signatures to monads in UniMath"),we described a category-theoretic construction of abstract syntax from a signature, mechanized in the UniMath library based on the Coq proof assistant.

    In the present work, we describe what was necessary to generalize that work to account for simply-typed languages. First, some definitions had to be generalized to account for the natural appearance of non-endofunctors in the simply-typed case. As it turns out, in many cases our mechanized results carried over to the generalized definitions without any code change. Second, an existing mechanized library on 𝜔-cocontinuous functors had to be extended by constructions and theorems necessary for constructing multi-sorted syntax. Third, the theoretical framework for the semantical signatures had to be generalized from a monoidal to a bicategorical setting, again to account for non-endofunctors arising in the typed case. This uses actions of endofunctors on functors with given source, and the corresponding notion of strong functors between actions, all formalized in UniMath using a recently developed library of bicategory theory. We explain what needed to be done to plug all of these ingredients together, modularly.

    The main result of our work is a general construction that, when fed with a signature for a simply-typed language, returns an implementation of that language together with suitable boilerplate code, in particular, a certified monadic substitution operation.

  • 13.
    Alam, Mahbub Ul
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Advancing Clinical Decision Support Using Machine Learning & the Internet of Medical Things: Enhancing COVID-19 & Early Sepsis Detection2024Doctoral thesis, comprehensive summary (Other academic)
    Abstract [en]

    This thesis presents a critical examination of the positive impact of Machine Learning (ML) and the Internet of Medical Things (IoMT) for advancing the Clinical Decision Support System (CDSS) in the context of COVID-19 and early sepsis detection.

    It emphasizes the transition towards patient-centric healthcare systems, which necessitate personalized and participatory care—a transition that could be facilitated by these emerging fields. The thesis accentuates how IoMT could serve as a robust platform for data aggregation, analysis, and transmission, which could empower healthcare providers to deliver more effective care. The COVID-19 pandemic has particularly stressed the importance of such patient-centric systems for remote patient monitoring and disease management.

    The integration of ML-driven CDSSs with IoMT is viewed as an extremely important step in healthcare systems that could offer real-time decision-making support and enhance patient health outcomes. The thesis investigates ML's capability to analyze complex medical datasets, identify patterns and correlations, and adapt to changing conditions, thereby enhancing its predictive capabilities. It specifically focuses on the development of IoMT-based CDSSs for COVID-19 and early sepsis detection, using advanced ML methods and medical data.

    Key issues addressed cover data annotation scarcity, data sparsity, and data heterogeneity, along with the aspects of security, privacy, and accessibility. The thesis also intends to enhance the interpretability of ML prediction model-based CDSSs. Ethical considerations are prioritized to ensure adherence to the highest standards.

    The thesis demonstrates the potential and efficacy of combining ML with IoMT to enhance CDSSs by emphasizing the importance of model interpretability, system compatibility, and the integration of multimodal medical data for an effective CDSS.

    Overall, this thesis makes a significant contribution to the fields of ML and IoMT in healthcare, featuring their combined potential to enhance CDSSs, particularly in the areas of COVID-19 and early sepsis detection.

    The thesis hopes to enhance understanding among medical stakeholders and acknowledges the need for continuous development in this sector.

    Download full text (pdf)
    Advancing Clinical Decision Support Using Machine Learning & the Internet of Medical Things
    Download (jpg)
    omslagsframsida
  • 14.
    Alam, Mahbub Ul
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Baldvinsson, Jón Rúnar
    Skatturinn (Iceland Revenue and Customs), Reykjavík, Iceland.
    Wang, Yuxia
    Qamcom Research and Technology, Stockholm, Sweden.
    Exploring LRP and Grad-CAM visualization to interpret multi-label-multi-class pathology prediction using chest radiography2022In: 2022 IEEE 35th International Symposium on Computer-Based Medical Systems (CBMS) / [ed] Linlin Shen; KC Santosh; Alejandro Rodríguez González, IEEE conference proceedings , 2022, p. 258-263Conference paper (Refereed)
    Abstract [en]

    The area of interpretable deep neural networks has received increased attention in recent years due to the need for transparency in various fields, including medicine, healthcare, stock market analysis, compliance with legislation, and law. Layer-wise Relevance Propagation (LRP) and Gradient-weighted Class Activation Mapping (Grad-CAM) are two widely used algorithms to interpret deep neural networks. In this work, we investigated the applicability of these two algorithms in the sensitive application area of interpreting chest radiography images. In order to get a more nuanced and balanced outcome, we use a multi-label classification-based dataset and analyze the model prediction by visualizing the outcome of LRP and Grad-CAM on the chest radiography images. The results show that LRP provides more granular heatmaps than Grad-CAM when applied to the CheXpert dataset classification model. We posit that this is due to the inherent construction difference of these algorithms (LRP is layer-wise accumulation, whereas Grad-CAM focuses primarily on the final sections in the model's architecture). Both can be useful for understanding the classification from a micro or macro level to get a superior and interpretable clinical decision support system.

  • 15.
    Alam, Mahbub Ul
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Hollmén, Jaakko
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Baldvinsson, Jón Rúnar
    Skatturinn (Iceland Revenue and Customs), Iceland.
    Rahmani, Rahim
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    SHAMSUL: Systematic Holistic Analysis to investigate Medical Significance Utilizing Local interpretability methods in deep learning for chest radiography pathology prediction2023In: Nordic Machine Intelligence, ISSN 2703-9196, Vol. 3, no 1, p. 27-47Article in journal (Refereed)
    Abstract [en]

    The interpretability of deep neural networks has become a subject of great interest within the medical and healthcare domain. This attention stems from concerns regarding transparency, legal and ethical considerations, and the medical significance of predictions generated by these deep neural networks in clinical decision support systems. To address this matter, our study delves into the application of four well-established interpretability methods: Local Interpretable Model-agnostic Explanations (LIME), Shapley Additive exPlanations (SHAP), Gradient-weighted Class Activation Mapping (Grad-CAM), and Layer-wise Relevance Propagation (LRP). Leveraging the approach of transfer learning with a multi-label-multi-class chest radiography dataset, we aim to interpret predictions pertaining to specific pathology classes. Our analysis encompasses both single-label and multi-label predictions, providing a comprehensive and unbiased assessment through quantitative and qualitative investigations, which are compared against human expert annotation. Notably, Grad-CAM demonstrates the most favorable performance in quantitative evaluation, while the LIME heatmap score segmentation visualization exhibits the highest level of medical significance. Our research underscores both the outcomes and the challenges faced in the holistic approach adopted for assessing these interpretability methods and suggests that a multimodal-based approach, incorporating diverse sources of information beyond chest radiography images, could offer additional insights for enhancing interpretability in the medical domain.

    Download full text (pdf)
    SHAMSUL
  • 16.
    Alam, Mahbub Ul
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Hollmén, Jaakko
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Rahmani Chianeh, Rahim
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    COVID-19 detection from thermal image and tabular medical data utilizing multi-modal machine learning2023In: 2023 IEEE 36th International Symposium on Computer-Based Medical Systems (CBMS): 22-24 June 2023 / [ed] João Rafael Almeida; Myra Spiliopoulou; Jose Alberto Benitez Andrades; Giuseppe Placidi; Alejandro Rodríguez González; Rosa Sicilia; Bridget Kane, 2023, p. 646-653Conference paper (Refereed)
    Abstract [en]

    COVID-19 is a viral infectious disease that has created a global pandemic, resulting in millions of deaths and disrupting the world order. Different machine learning and deep learning approaches were considered to detect it utilizing different medical data. Thermal imaging is a promising option for detecting COVID-19 as it is low-cost, non-invasive, and can be maintained remotely. This work explores the COVID-19 detection issue using the thermal image and associated tabular medical data obtained from a publicly available dataset. We incorporate a multi-modal machine learning approach where we investigate the different combinations of medical and data type modalities to get an improved result. We use different machine learning and deep learning methods, namely random forests, Extreme Gradient Boosting (XGBoost), Multilayer Perceptron (MLP), and Convolutional Neural Network (CNN). Overall multi-modal results outperform any single modalities, and it is observed that the thermal image is a crucial factor in achieving it. XGBoost provided the best result with the area under the receiver operating characteristic curve (AUROC) score of 0.91 and the area under the precision-recall curve (AUPRC) score of 0.81. We also report the average of leave-one-positive-instance-out cross- validation evaluation scores. This average score is consistent with the test evaluation score for random forests and XGBoost methods. Our results suggest that utilizing thermal image with associated tabular medical data could be a viable option to detect COVID-19, and it should be explored further to create and test a real-time, secure, private, and remote COVID-19 detection application in the future.

  • 17.
    Alam, Mahbub Ul
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Rahmani, Rahim
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Cognitive Internet of Medical Things Architecture for Decision Support Tool to Detect Early Sepsis Using Deep Learning2021In: Biomedical Engineering Systems and Technologies: BIOSTEC 2020 / [ed] Xuesong Ye, Filipe Soares, Elisabetta De Maria, Pedro Gómez Vilda, Federico Cabitza, Ana Fred, Hugo Gamboa, Cham: Springer, 2021, Vol. 1400, p. 366-384Chapter in book (Refereed)
    Abstract [en]

    The internet of medical things (IoMT) is a relatively new territory for the internet of things (IoT) platforms where we can obtain a significant amount of potential benefits with the combination of cognitive computing. Effective utilization of the healthcare data is the critical factor in achieving such potential, which can be a significant challenge as the medical data is extraordinarily heterogeneous and spread across different devices with different degrees of importance and authority. To address this issue, in this paper, we introduce a cognitive internet of medical things architecture with a use case of early sepsis detection using electronic health records. We discuss the various aspects of IoMT architecture. Based on the discussion, we posit that the proposed architecture could improve the overall performance and usability in the IoMT platforms in particular for different IoMT based services and applications. The use of an RNN-LSTM network for early prediction of sepsis according to Sepsis-3 criteria is evaluated with the empirical investigation using six different time window sizes. The best result is obtained from a model using a four-hour window with the assumption that data is missing-not-at-random. It is observed that when learning from heterogeneous sequences of sparse medical data for early prediction of sepsis, the size of the time window has a considerable impact on predictive performance.

  • 18.
    Alam, Mahbub Ul
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Rahmani, Rahim
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Federated Semi-Supervised Multi-Task Learning to Detect COVID-19 and Lungs Segmentation Marking Using Chest Radiography Images and Raspberry Pi Devices: An Internet of Medical Things Application2021In: Sensors, E-ISSN 1424-8220, Vol. 21, no 15, article id 5025Article in journal (Refereed)
    Abstract [en]

    Internet of Medical Things (IoMT) provides an excellent opportunity to investigate better automatic medical decision support tools with the effective integration of various medical equipment and associated data. This study explores two such medical decision-making tasks, namely COVID-19 detection and lung area segmentation detection, using chest radiography images. We also explore different cutting-edge machine learning techniques, such as federated learning, semi-supervised learning, transfer learning, and multi-task learning to explore the issue. To analyze the applicability of computationally less capable edge devices in the IoMT system, we report the results using Raspberry Pi devices as accuracy, precision, recall, Fscore for COVID-19 detection, and average dice score for lung segmentation detection tasks. We also publish the results obtained through server-centric simulation for comparison. The results show that Raspberry Pi-centric devices provide better performance in lung segmentation detection, and server-centric experiments provide better results in COVID-19 detection. We also discuss the IoMT application-centric settings, utilizing medical data and decision support systems, and posit that such a system could benefit all the stakeholders in the IoMT domain.

    Download full text (pdf)
    Federated Semi-Supervised Multi-Task Learning to Detect COVID-19 and Lungs Segmentation Marking Using Chest Radiography Images and Raspberry Pi Devices: An Internet of Medical Things Application
  • 19.
    Allaart, Corinne
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences. Karolinska Institute, Sweden.
    Mondrejevski, Lena
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences. Karolinska Institute, Sweden.
    Papapetrou, Panagiotis
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    FISUL: A Framework for Detecting Adverse Drug Events from Heterogeneous Medical Sources Using Feature Importance2019In: Artificial Intelligence Applications and Innovations: Proceedings / [ed] John MacIntyre, Ilias Maglogiannis, Lazaros Iliadis, Elias Pimenidis, Springer, 2019, p. 139-151Conference paper (Refereed)
    Abstract [en]

    Adverse drug events (ADEs) are considered to be highly important and critical conditions, while accounting for around 3.7% of hospital admissions all over the world. Several studies have applied predictive models for ADE detection; nonetheless, only a restricted number and type of features has been used. In the paper, we propose a framework for identifying ADEs in medical records, by first applying the Boruta feature importance criterion, and then using the top-ranked features for building a predictive model as well as for clustering. We provide an experimental evaluation on the MIMIC-III database by considering 7 types of ADEs illustrating the benefit of the Boruta criterion for the task of ADE detection.

  • 20.
    Almeida, João Paolo A.
    et al.
    Federal University of Espírito Santo, Vitória, Brazil.
    Borbinha, JoséUniversidade de Lisboa, Lisbon, Portugal.Guizzardi, GiancarloStockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.Link, SebastianUniversity of Auckland, Auckland, New Zealand.Zdravkovic, JelenaStockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Conceptual Modeling: 42nd International Conference, ER 2023, Lisbon, Portugal, November 6–9, 2023, Proceedings2023Collection (editor) (Other academic)
    Abstract [en]

    This book constitutes the refereed proceedings of the 42nd International Conference on Conceptual Modeling, ER 2023, held in Lisbon, Portugal, during November 6-9, 2023. The 21 full papers were carefully reviewed and selected from 121 submissions. Additionally, the book contains 4 keynote speeches and 3 tutorials, and one invited paper corresponding to one of the keynote speeches.

  • 21. Alpcan, Tansu
    et al.
    Everitt, Tom
    Stockholm University, Faculty of Science, Department of Mathematics.
    Hutter, Marcus
    Can we measure the difficulty of an optimization problem?2014Conference paper (Other academic)
    Abstract [en]

    Can we measure the difficulty of an optimization problem? Although optimization plays a crucial role in modernscience and technology, a formal framework that puts problemsand solution algorithms into a broader context has not beenestablished. This paper presents a conceptual approach which gives a positive answer to the question for a broad class of optimization problems. Adopting an information and computational perspective, the proposed framework builds upon Shannon and algorithmic information theories. As a starting point, a concrete model and definition of optimization problems is provided. Then, a formal definition of optimization difficulty is introduced which builds upon algorithmic information theory. Following an initial analysis, lower and upper bounds on optimization difficulty are established. One of the upper-bounds is closely related to Shannon information theory and black-box optimization. Finally, various computational issues and future research directions are discussed.

  • 22.
    Althini, Vera
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Larsson, Rebecka
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Ett designverktygs roll i samarbetet mellan designers och utvecklare: En kvalitativ studie om hur Figma används i samarbeten mellan designers och utvecklare2023Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
    Abstract [en]

    In the field of software development, collaboration between designers and other technical stakeholders such as developers is common. During the collaboration, one occurring process is the so-called handoff, in which designers hand over a design to developers in order to implement the design into code. Today, designers and developers can often work together in an iterative fashion. Despite this, challenges may occur during the collaboration, such as when trying to transform design to code. Other breakdowns may also occur during the collaboration.

    One tool, that is used for design but can also offer exporting of elements and transforming to code, is Figma. The tool is described as being convenient for collaboration, such as designers and developers among other groups. According to Figma, using the tool may transform a designer-developer handoff into a “handshake”. In this thesis, the role of Figma in the designers-developer collaboration and its usage is studied. The goal of the study is to provide an answer to how Figma is used in the collaboration between designers and developers at the end of the design process and during the construction process. The construction process describes the process where the implementation is done, according to the authors' own understanding.

    Semi-structured digital interviews with three designers and four developers from companies in Sweden were conducted in this study. The majority of the participants worked at different companies. Affinity diagram were used to analyze the collected data.

    The results indicated that designers and developers usually use Figma throughout the design process, but also in the intersection of the design- and construction process; designers create guidelines in the design system and use comments to share information with developers. The developers review and approve designers’ creations before inserting them into the design system, or they just create the components themselves. Developers can also use the design system as a foundation for coding. Another interesting find was that other aspects outside of Figma, such as human factors, could impact the collaboration as well. Both designers and developers use Figma to communicate by text and through visuals, such as via the comment feature, prototypes and design systems.

    After discussing the results, it became clear the Figma is used during handoffs but not exclusively so. The tool can also be used for more than just creating interfaces. In many ways Figma can support designers and developers in speaking the same language, such as by having similar names to the ones found in code. On the other hand, the participants saw room for improvement in this area since some features and names in Figma differ from code. Another frequent challenge when collaborating in Figma was the synchronization between designers’ and developers’ representations, where representations are the things that designers and developers create respectively. In other words, designers and developers described challenges in picking up the changes made in Figma or in the development environment, and maintaining a single source of truth.

    The conclusion is that Figma is used in a collaborative design process to communicate the behaviour and visual aspects of the design. During the construction process, Figma is used to review design artefacts and inspect the design system in Figma, which acts as foundations for implementation.

    Download full text (pdf)
    FULLTEXT01
  • 23.
    Al-Towhi, Khalil
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    The ideal use of NFT in Metaverse - A Systematic literature review2023Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    It has become possible to say that the metaverse is a great opportunity for investment and trade, as it provides massive financial returns. The metaverse is the next evolution in social connection and the successor to the mobile internet. Non-fungible tokens represent the ownership of unique items in the metaverse and allow the creator to tokenize things like art, real estate, and collectibles. Trading NFT in the metaverse faces challenges, including security, fraud, and scams. Those challenges have negatively affected the stability of this market. “How can NFT trading in metaverse be improved?” is the main question of this thesis to overcome the challenges. The author performed a systematic literature review to survey and explore the possibility of using technologies to reach the ideal use of NFT in the metaverse. The systematic literature review will guide the researcher to gain more information to evaluate it in the research area. Furthermore, Pointing and identifying the gaps and knowledge needed between the research elements. Three main challenges are presented (identity verification, fraud, and ownership) in areas in which technologies that can provide (flexibility, reliability, accuracy, and performance) can apply. Three dominant solutions, smart contracts, oracle nodes, and blockchain are the study and analysis results to realize the research question and identified problem. The elected technologies show an ability to address challenges in different ways and thus maintain the security and effectiveness of trading operations. Also, the result section mentions other solutions not counting on the dominant solutions. Open issues which provide a ground for future research with practical implementations are also discussed.

    Download full text (pdf)
    FULLTEXT01
  • 24.
    Altschuler, Robert Henry
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Analyzing Motivation generated from Data Physicalizations2023Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    This thesis investigates the impact of data physicalizations in the form of an exercise data tracker, on user motivation compared to traditional exercise tracking devices (ETDs) to and their effect on exercise motivation. By employing a case study with diverse participants, this research evaluates whether the three-dimensional nature of data physicalizations significantly influences user motivation or if similar results can be obtained using traditional solutions or devices. Through a pre-interview Leisure-Time Exercise Questionnaire, qualitative interviews, and thematic analysis, six significant themes emerged, presence, emotional connection, data clarity, metric variety, aesthetics, and novelty. The results suggest that data physicalizations have the potential to motivate users, but are currently at a disadvantage compared to traditional ETDs due to limitations in data precision and variety. The study also highlighted the design challenges in developing effective data physicalizations. Future research should explore design features highlighted in this study that could potentially maximize motivation.

    Download full text (pdf)
    FULLTEXT01
  • 25.
    Amneryd, Ida
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Aldafae, Anwar
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Designing for All Gamers2023Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
    Abstract [en]

    In recent years, playing games has become a popular activity for everything from socialising purposes to educating purposes, with billions of players worldwide. However, being able to play games is not evident for people with disabilities. The gaming industry is excluding masses of people by not creating enough accessible games. Indie gaming is a part of the gaming industry that has grown immensely during the last decades. Indie game companies are often smaller studios with independent publications, finances and creativity. Almost everyone has the tools nowadays for making games. Indie game companies do not usually have the resources for engaging as much in accessibility as bigger companies do. Since they are distributing a big share of the games on the market today it is necessary to explore their possibilities of creating more accessible games despite their resource constraints. The research question for this study is “What are the game development challenges that Swedish indie game companies face when implementing accessible design practices, and how can they overcome them?”. This was investigated through a survey study where the researchers conducted semi-structured interviews with employees at seven different indie game companies in Sweden. The results revealed perceived challenges in the tools that the investigated companies utilised, such as challenges in accessibility communication, resource constraints and how they aim to gather player feedback. Other challenges were identified in how the company operates, more specifically in their work tasks and planning of accessibility. Results also contributed to an understanding of how the investigated companies work with accessibility, through how they prioritise tasks and how they receive accessibility information. Possible solutions on how they can overcome challenges were identified, both on an individual and company level, as well as on an industry level for the indie gaming market. The results provided insights into the factors impeding indie game companies and developers from designing for accessibility as well as possible solutions to these challenges. These findings contribute to the understanding of the challenges with implementing accessible design practices and have added a nuanced perspective on the challenges faced by indie game developers and suggest potential directions for future investigation, both in research and in the indie game industry.

    Download full text (pdf)
    FULLTEXT01
  • 26.
    Anderson, Ian
    et al.
    University of Bristol.
    Maitland, Julie
    University of Glasgow.
    Sherwood, Scott
    University of Glasgow.
    Barkhuus, Louise
    Institutionen för data- och systemvetenskap, ACT Agera i kommunikation med teknik.
    Chalmers, Matthew
    University of Glasgow.
    Hall, Malcolm
    University of Glasgow.
    Brown, Barry
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Muller, Henk
    University of Bristol.
    Shakra: tracking and sharing daily activity levels with unaugmented mobile phones2007In: Mobile Networks and Applications, ISSN 1383-469X, E-ISSN 1572-8153, Vol. 12, no 2-3, p. 185-199Article in journal (Refereed)
    Abstract [en]

    This paper explores the potential for use of an unaugmented commodity technology—the mobile phone— as a health promotion tool. We describe a prototype application that tracks the daily exercise activities of people, using an Artificial Neural Network (ANN) to analyse GSM cell signal strength and visibility to estimate a user’s movement. In a short-term study of the prototype that shared activity information amongst groups of friends, we found that awareness encouraged reflection on, and increased motivation for, daily activity. The study raised concerns regarding the reliability of ANN-facilitated activity detection in the ‘real world’. We describe some of the details of the pilot study and introduce a promising new approach to activity detection that has been developed in response to some of the issues raised by the pilot study, involving Hidden Markov Models (HMM), task modelling and unsupervised calibration. We conclude with our intended plans to develop the system further in order to carry out a longer-term clinical trial.

  • 27.
    Andersson, Birger
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Kabilan, Vandana
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    García Lozano, Marianela
    Mojtahed, Vahid
    Svan, Pernilla
    Konceptuell Modellering inom det Svenska Försvaret - DCMF2006Report (Other academic)
  • 28.
    Andersson, Birger
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Mojtahed, Vahid
    Ulriksson, Jenny
    Eklöf, Martin
    Svan, Pernilla
    Konceptdemonstrator. Ett verktyg för att underlätta och påskynda Försvarets förändringsarbete2006Report (Other academic)
  • 29.
    Andersson, Jonathan
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Jedeur-Palmgren, Arvid
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Exploring the intersection of deceptive designs and user perceptions in the data economy2023Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
    Abstract [en]

    When users in the EU visits a website, they will face a cookie consent banner because of the current regulations. A framework called “notice-and-consent” is in place to inform users about cookies and their intentions and to receive consent from the user. Websites have been shown to apply deceptive designs in their cookie notice in order to trick the user into consenting to more cookies than they are aware of, and these deceptive designs are referred to as dark patterns.

    The framework used today for regulating data collection has met some criticism. It enables businesses to do what they choose with the information they gather as long as they inform the user of their intention. In addition, previous research has shown that businesses can apply dark patterns on the cookie consent notice to gain an economic advantage by making users consent to more cookies than they understand when faced with the decision.

    This study investigates how websites targeting visitors in Sweden are using dark patterns in their cookie consent notices and how this affects the online experience. Further, this study sought to investigate the end-user’s perception and experience of the data collection climate online by answering the following question: Which dark patterns can be found in the most visited websites in Sweden and how do these affect the users' online experience and perceptions?

    The research strategy used in this study was a survey strategy. The data collection method was a manual data collection on websites targeting a visitor from Sweden followed by a qualitative online questionnaire, and the data analysis method was a thematic analysis. The study resulted in finding out that 82% of the websites used at least one dark pattern in their cookie and consent interface. The most utilised patterns were Interface interference and Obstruction, found on 62% and 58% of the websites. Most respondents showed signs of irritation or fatigue towards the cookie consent notices or an overall negative perception of the current data collection climate. The respondents also described experiences with the most utilised dark patterns. Further, this study concludes that dark patterns are used as extensively in Sweden as it is in other domains. The regulation that is supposed to protect users might not be the immediate problem but can sometimes be seen as non-existent. A lack of enforcement leads to a market that is free to function on its own. Therefore the responsibility should be removed from the user and instead regulate what is allowed to do with personal data.

    Download full text (pdf)
    FULLTEXT01
  • 30.
    Andersson, Niklas
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    The effect of the IT/OT gap on the NIS 2 implementation2023Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    Cyber attacks are steadily increasing, and their impact is becoming more significant. To combat this, the European Union has created directives to enhance the cyber security in critical services in the Union, one example being the NIS 2 directive. The directive comes into force during the fourth industrial revolution, where the Operational Technology (OT) is connected to the Information Technology (IT). This creates new vulnerabilities in the OT environments since they can now suffer from cyber attacks. The historical ways of securing OT and IT environments differ, which has caused what is called the IT/OT gap now that they are converging. In order to implement the NIS 2 directive and to enhance the cyber security of the entire organization, the IT/OT gap needs to be minimized.

    The problem this study then aims to investigate is how the effects of the IT/OT gap can be reduced in the implementation of the NIS 2 directive. This was done by answering the research question: To what extent is the IT/OT gap a challenge for the implementation of the NIS 2 directive in Sweden?

    The sub-question: In what areas is the IT/OT gap problematic for the implementation of the NIS 2 directive in Sweden?

    To gain an answer to the research question semi-structured interviews were conducted with respondents with knowledge in IT and OT security as well as the NIS 2 directive. The interviews were transcribed and analyzed using a thematic analysis.

    The thematic analysis resulted in 6 themes, Need for technical solutions, Lacking resources, Differences in security culture, Lack of cooperation, Supervisory authority and Standards, and six subthemes. The result showed that the IT/OT gap is a challenge for the implementation of the NIS 2 directive in a varying degree depending on the company. Further, it was shown that the IT/OT gap is most likely a problem in the areas regarding the supervisory authority, lacking resources, and cooperation.

    To comply with the directive and, more importantly, raise the level of cyber security, organizations and companies must handle all their risk in both IT and OT environments. The OT and IT personnel will need to talk to each other and collaborate to do it, and that might be a significant first step to minimizing the IT/OT gap in the long term.

    Download full text (pdf)
    FULLTEXT01
  • 31.
    Angelov, Krasimir
    et al.
    University of Gothenburg, Sweden.
    Liefke, KristinaGoethe University, Germany.Loukanova, RoussankaStockholm University, Faculty of Science, Department of Mathematics. Stockholm University, Faculty of Humanities, Department of Philosophy.Moortgat, MichaelUtrecht University, The Netherlands.Tojo, SatoshiSchool of Information Science, JAIST, Japan.
    Proceedings of the Symposium on Logic and Algorithms in Computational Linguistics 2018 (LACompLing2018)2018Conference proceedings (editor) (Refereed)
    Abstract [en]

    Computational linguistics studies natural language in its various manifestations from a computational point of view, both on the theoretical level (modeling grammar modules dealing with natural language form and meaning, and the relation between these two) and on the practical level (developing applications for language and speech technology). Right from the start in the 1950ties, there have been strong links with computer science, logic, and many areas of mathematics - one can think of Chomsky's contributions to the theory of formal languages and automata, or Lambek's logical modeling of natural language syntax. The symposium on Logic and Algorithms in Computational Linguistics 2018 (LACompLing2018) assesses the place of logic, mathematics, and computer science in present day computational linguistics. It intends to be a forum for presenting new results as well as work in progress.

    Download full text (pdf)
    fulltext
  • 32. Angleby, Helen
    et al.
    Oskarsson, Mattias
    Pang, Junfeng
    Zhang, Ya-ping
    Leitner, Thomas
    Braham, Caitlyn
    Arvestad, Lars
    Stockholm University, Faculty of Science, Numerical Analysis and Computer Science (NADA). Stockholm University, Science for Life Laboratory (SciLifeLab).
    Lundeberg, Joakim
    Webb, Kristen M.
    Savolainen, Peter
    Forensic Informativity of similar to 3000bp of Coding Sequence of Domestic Dog mtDNA2014In: Journal of Forensic Sciences, ISSN 0022-1198, E-ISSN 1556-4029, Vol. 59, no 4, p. 898-908Article in journal (Refereed)
    Abstract [en]

    The discriminatory power of the noncoding control region (CR) of domestic dog mitochondrial DNA alone is relatively low. The extent to which the discriminatory power could be increased by analyzing additional highly variable coding regions of the mitochondrial genome (mtGenome) was therefore investigated. Genetic variability across the mtGenome was evaluated by phylogenetic analysis, and the three most variable similar to 1kb coding regions identified. We then sampled 100 Swedish dogs to represent breeds in accordance with their frequency in the Swedish population. A previously published dataset of 59 dog mtGenomes collected in the United States was also analyzed. Inclusion of the three coding regions increased the exclusion capacity considerably for the Swedish sample, from 0.920 for the CR alone to 0.964 for all four regions. The number of mtDNA types among all 159 dogs increased from 41 to 72, the four most frequent CR haplotypes being resolved into 22 different haplotypes.

  • 33. Antosz, Patrycja
    et al.
    Birks, Dan
    Edmonds, Bruce
    Heppenstall, Alison
    Meyer, Ruth
    Polhill, J. Gareth
    O'Sullivan, David
    Wijermans, Nanda
    Stockholm University, Faculty of Science, Stockholm Resilience Centre. Institute for Future Studies, Sweden.
    What do you want theory for? - A pragmatic analysis of the roles of theory in agent-based modelling2023In: Environmental Modelling & Software, ISSN 1364-8152, E-ISSN 1873-6726, Vol. 168, article id 105802Article in journal (Refereed)
    Abstract [en]

    There has been some discussion about agent-based modelling (ABM) and theory, particularly how ABM might facilitate theory building. However, there is confusion about the different ways they could relate and some scepticism as to whether theory is needed if one has an ABM. This paper distinguishes some of the different ways that the term “theory” is used in ABM papers in three important ABM journals: Environmental Modelling & SoftwareComputers, Environment and Urban Systems and the Journal of Artificial Societies and Social Simulation. Apart from the simple-minded identification of theory with mathematics, we distinguish nine different ways that theory and ABM relate. This analysis is situated with respect to some of the expectations and philosophical background behind the idea of “theory”. The paper concludes with some ways in which theory and ABM could work better together, some possible ways forward and suggests that a more cautious approach to generalisation might be more appropriate.

  • 34.
    Apostolopoulos, Stavros
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Perceptions and Adoption of Cryptocurrencies in the Aftermath of the Greek Financial Crisis. A Study on the Region of Eastern Macedonia and Thrace2023Independent thesis Advanced level (degree of Master (Two Years)), 20 credits / 30 HE creditsStudent thesis
    Abstract [en]

    This research study investigates the potential of cryptocurrencies in alleviating the consequences of the financial crisis and identifying new avenues for economic development in Eastern Macedonia and Thrace, Greece. It is motivated by the significant economic downturn experienced by Greece following the 2008 financial crisis, which led to a risk of bankruptcy and a drastic reduction in GDP per capita, with the consecutive implementation of capital controls further disrupting the financial transactions in the country. The region under study is documented for its low economic development level and GDP per capita compared to the national and EU averages. Through a mixed-methods approach, including a questionnaire-based survey and data analysis, this study explores the perceptions and behaviors of consumers and business owners in that region regarding cryptocurrencies. Furthermore, the survey employs closed-ended questions delivered through an internet-based platform while using probability and non-probability sampling techniques to target consumers and business owners. Consequently, the collected data are examined through the scopes of descriptive and deductive analyses with the use of SPSS software, with the findings of this research aiming to shed light on the role of cryptocurrencies as a means to mitigate the impact of the financial crisis and stimulate economic activity in Eastern Macedonia and Thrace. The findings revealed that while participants did not believe that introducing a parallel digital currency would improve the Greek economy, consumers affected by the crisis showed eagerness to invest and transact in cryptocurrencies. Business owners, on the other hand, were hesitant to view cryptocurrencies as long-term assets and did not believe in their capacity to transform the region’s economy. The study's outcomes contribute to the growing body of knowledge on cryptocurrencies' adoption and potential benefits in regions facing economic challenges.

    Download full text (pdf)
    FULLTEXT01
  • 35.
    Aranda, Laura
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Popov, Oliver
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    A Conceptual Model of an Intelligent Platform for Security Risk Assessment in SMEs2019In: 2019 IEEE 13th International Conference on Application of Information and Communication Technologies (AICT): Conference Proceedings, IEEE, 2019, p. 181-188Conference paper (Refereed)
    Abstract [en]

    SMEs are increasingly targeted by cyberattacks and usually less in control of their Information Security Management System than larger organizations due to a lack of resources. Risk assessment can help them to determine which changes are needed bearing in mind their constraints. However, common frameworks for risk assessments are more suitable for large organizations. Some of them have been designed specifically for SMEs but still target an audience of information security experts and are considered as time-consuming by SMEs. This article aims at tackling those issues by introducing a conceptual model of an Intelligent Platform for supporting SMEs in security risk assessment process. The design research method was used to develop a model taking into account the inputs from relevant stakeholders collected via interviews. The model was validated and improved with case studies where quick security risk assessments in three different SMEs have been performed following the activities that the proposed model is supposed to perform.

  • 36.
    Araújo, Marco
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Ekenberg, Love
    International Institute for Applied Systems Analysis (IIASA), Austria.
    Confraria, João
    School of Business & Economics, Catholic University, Portugal.
    Satellite backhaul for macro-cells, as an alternative to optical fibre, to close the digital divide2019In: 2019 IEEE Wireless Communications and Networking Conference (WCNC), IEEE, 2019Conference paper (Refereed)
    Abstract [en]

    The lack of broadband access causes a serious risk of social exclusion, by preventing citizens from full social and economic participation in the society. To avoid this risk, the concept of subsidized rural networks was developed by the European Commission, in which an operator builds, maintains and operates a network (usually an open network) capable of providing at least a 100 Mbps connection to the subscribers; deployed in low density regions being publicly subsidized when unprofitable. In this article, we suggest a methodology to measure a realistic value for the average broadband used per subscriber at busy hour. We also present a simulation model for the backhaul infrastructure costs for very fast networks in rural areas to cover the last, and more expensive, 5% of the population, while comparing optical fibre with satellite for the middle mile from an economical and financial perspective.

  • 37. Arrighi, Emmanuel
    et al.
    Fernau, Henning
    De Oliveira, Mateus
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Wolf, Petra
    Order Reconfiguration under Width Constraints2023In: Journal of Graph Algorithms and Applications, E-ISSN 1526-1719, Vol. 27, no 6, p. 409-431Article in journal (Refereed)
    Abstract [en]

    In this work, we consider the following order reconfiguration problem: Given a graph G together with linear orders ω and ω′ of the vertices of G, can one transform ω into ω′ by a sequence of swaps of adjacent elements in such a way that, at each time step, the resulting linear order has cutwidth (pathwidth) at most k? We show that this problem always has an affirmative answer when the input linear orders ω and ω′ have cutwidth (pathwidth) of at most k/2. This result also holds in a weighted setting. Using this result, we establish a connection between two apparently unrelated problems: the reachability problem for two-letter string rewriting systems and the graph isomorphism problem for graphs of bounded cutwidth. This opens an avenue for the study of the famous graph isomorphism problem using techniques from term rewriting theory.

  • 38.
    Arrighi, Emmanuel
    et al.
    University of Bergen.
    Fernau, Henning
    University of Trier.
    De Oliveira Oliveira, Mateus
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences. University of Bergen .
    Wolf, Petra
    University of Bergen; University of Trier.
    Synchronization and Diversity of Solutions2023In: Thirty-Seventh AAAI Conference on Artificial Intelligence: AAAI-23 Technical Tracks 10 / [ed] Brian Williams, Yiling Chen, Jennifer Neville, 2023, Vol. 37(10), p. 11516-11524Conference paper (Refereed)
    Abstract [en]

    A central computational problem in the realm of automata theory is the problem of determining whether a finite automaton A has a synchronizing word. This problem has found applications in a variety of subfields of artificial intelligence, including planning, robotics, and multi-agent systems. In this work, we study this problem within the framework of diversity of solutions, an up-and-coming trend in the field of artificial intelligence where the goal is to compute a set of solutions that are sufficiently distinct from one another. We define a notion of diversity of solutions that is suitable for contexts were solutions are strings that may have distinct lengths. Using our notion of diversity, we show that for each fixed r ∈ N, each fixed finite automaton A, and each finite automaton B given at the input, the problem of determining the existence of a diverse set {w1,w2, . . . ,wr} ⊆ L(B) of words that are synchronizing for A can be solved in polynomial time. Finally, we generalize this result to the realm of conformant planning, where the goal is to devise plans that achieve a goal irrespectively of initial conditions and of nondeterminism that may occur during their execution.

  • 39.
    Asplund, Einar
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Sandell, Johan
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Comparison of graph databases and relational databases performance2023Independent thesis Basic level (degree of Bachelor), 10 credits / 15 HE creditsStudent thesis
    Abstract [en]

    There has been a change of paradigm in which way information is being produced, processed, and consumed as a result of social media. While planning to store the data, it is important to choose a suitable database for the type of data, as unsuitable storage and analysis can have a noticeable impact on the system’s energy consumption. Additionally, effectively analyzing data is essential because deficient data analysis on a large dataset can lead to repercussions due to unsound decisions and inadequate planning. In recent years, an increasing amount of organizations have provided services that cannot be anymore achieved efficiently using relational databases. An alternative data storage is graph databases, which is a powerful solution for storing and searching for relationship-dense data. The research question that the thesis aims to answer is, how do state-of-the-art-graph database and relational database technologies compare with each other from a performance perspective in terms of time taken to query, CPU usage, memory usage, power usage, and temperature of the server?

    To answer the research question, an experimental study using analysis of variance will be performed. One relational database, MySQL, and two graph databases, ArangoDB and Neo4j, will be compared using a benchmark. The benchmark used is Novabench. The results from the post-hoc, KruskalWallis, and analysis of variances show that there are significant differences between the database technologies. This means the null hypothesis, that there is no significant difference, is rejected, and the alternative hypothesis, that there is a significant difference in performance between the database technologies in the aspects of Time to Query, Central Processing Unit usage, Memory usage, Average Energy usage, and temperature holds. In conclusion, the research question was answered. The study shows that Neo4j was the fastest at executing queries, followed by MySQL, and in last place ArangoDB. The results also showed that MySQL was more demanding on memory usage than the other database technologies.

    Download full text (pdf)
    FULLTEXT01
  • 40. Austrin, Per
    et al.
    Manokaran, Rajsekar
    Wenner, Cenny
    Stockholm University, Faculty of Science, Numerical Analysis and Computer Science (NADA). KTH - Royal Institute of Technology, Sweden.
    On the NP-Hardness of Approximating Ordering-Constraint Satisfaction Problems2015In: Theory of Computing, E-ISSN 1557-2862, Vol. 11, article id 10Article in journal (Refereed)
    Abstract [en]

    We show improved NP-hardness of approximating Ordering Constraint Satis-faction Problems (OCSPs). For the two most well-studied OCSPs, Maximum Acyclic Subgraph and Maximum Betweenness, we prove NP-hard approximation factors of 14/15+ε and 1/2+ε. When it is hard to approximate an OCSP by a constant better than takinga uniformly-at-random ordering, then the OCSP is said to be approximation resistant. We show that the Maximum Non-Betweenness Problem is approximation resistant and that there are width-m approximation-resistant OCSPs accepting only a fraction 1/(m/2)! of assignments. These results provide the first examples of approximation-resistant OCSPs only to P != NP.

    Download full text (pdf)
    fulltext
  • 41.
    Ayele, Workneh Y.
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Juell-Skielse, Gustaf
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    A Process Model for Generating and Evaluating Ideas: The Use of Machine Learning and Visual Analytics to Support Idea Mining2020In: Electronic Government and the Information Systems Perspective: 9th International Conference, EGOVIS 2020, Bratislava, Slovakia, September 14–17, 2020, Proceedings / [ed] Andrea Kő, Enrico Francesconi, Gabriele Kotsis, A Min Tjoa, Ismail Khalil, Springer, 2020, p. 189-203Conference paper (Refereed)
    Abstract [en]

    The significance and possibilities of idea generation and evaluation are increasing due to the increasing demands for digital innovation and the abundance of textual data. Textual data such as social media, publications, patents, documents, etc. are used to generate ideas, yet manual analysis is affected by bias and subjectivity. Machine learning and visual analytics tools could be used to support idea generation and evaluation, referred to as idea mining, to unlock the potential of voluminous textual data. Idea mining is applied to support the extraction of useful information from textual data. However, existing literature merely focuses on the outcome and overlooks structuring and standardizing the process itself. In this paper, to support idea mining, we designed a model following design science research, which overlaps with the Cross-Industry-Standard-Process for Data Mining (CRISP-DM) process and adapts well-established models for technology scouting. The first layer presents and business layer, where tasks performed by technology scouts, incubators, accelerators, consultants, and contest managers are detailed. The second layer presents the technical layer where tasks performed by data scientists, data engineers, and similar experts are detailed overlapping with CRISP-DM. For future research, we suggest an ex-post evaluation and customization of the model to other techniques of idea mining.

    Download full text (pdf)
    fulltext
  • 42.
    Ayele, Workneh Y.
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Juell-Skielse, Gustaf
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Eliciting Evolving Topics, Trends and Foresight about Self-driving Cars Using Dynamic Topic Modeling2020In: Advances in Information and Communication: Proceedings of the 2020 Future of Information and Communication Conference (FICC), Volume 1 / [ed] Kohei Arai; Supriya Kapoor; Rahul Bhatia, Cham: Springer, 2020, p. 488-509Conference paper (Refereed)
    Abstract [en]

    Self-driving technology is part of smart city ecosystems, and it touches a broader research domain. There are advantages associated with using this technology, such as improved quality of life, reduced pollution, and reduced fuel cost to name a few. However, there are emerging concerns, such as the impact of this technology on transportation systems, safety, trust, affordability, control, etc. Furthermore, self-driving cars depend on highly complex algorithms. The purpose of this research is to identify research agendas and innovative ideas using unsupervised machine learning, dynamic topic modeling, and to identify the evolution of topics and emerging trends. The identified trends can be used to guide academia, innovation intermediaries, R&D centers, and the auto industry in eliciting and evaluating ideas. The research agendas and innovative ideas identified are related to intelligent transportation, computer vision, control and safety, sensor design and use, machine learning and algorithms, navigation, and human-driver interaction. The result of this study shows that trending terms are safety, trust, transportation system (traffic, modeling traffic, parking, roads, power utilization, the buzzword smart, shared resources), design for the disabled, steering and control, requirement handling, machine learning, LIDAR (Light Detection And Ranging) sensor, real-time 3D image processing, navigation, and others. 

  • 43.
    Ayele, Workneh Y.
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Juell-Skielse, Gustaf
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Unveiling Topics from Scientific Literature on the Subject of Self-driving Cars using Latent Dirichlet Allocation2018In: 2018 IEEE 9th Annual Information Technology, Electronics and Mobile Communication Conference (IEMCON): Proceedings, IEEE, 2018, p. 1113-1119Conference paper (Refereed)
    Abstract [en]

    Self-driving cars are becoming popular topics in academia. Consumers of self-driving cars and vehicles have different concerns, for example, safety and security, to name a few. Also, the public sector has interests in self-driving cars such as amending policies to enable the management of self-driving vehicles in cities, urban planning, traffic management and, etc. In this paper, more than 2700 corpus are extracted from literature from several subject areas to identify latent (hidden) topics of self-driving cars. Latent Dirichlet Allocation (LDA) is used for topic identification. The result of this study shows that topics identified are valid research areas such as urban planning, driver car (computer) interaction, self-driving control and system design, ethics in self-driving cars, safety and risk assessment, training dataset quality and machine learning in self-driving cars are among the topics identified. Furthermore, the network visualization of association graph of terms shows that the most frequently discussed concepts reveal that control of self-driving cars is based on algorithms, data, design, method, and model. The methods used in this study and the results can be used as decision tools, if carefully applied, in diverse disciplines that are disrupted by the introduction of self-driving cars. For future study, we plan to extend this study with a larger dataset and other data mining techniques.

  • 44.
    Ayele, Workneh Yilma
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    A toolbox for idea generation and evaluation: Machine learning, data-driven, and contest-driven approaches to support idea generation2022Doctoral thesis, comprehensive summary (Other academic)
    Abstract [en]

    Ideas are sources of creativity and innovation, and there is an increasing demand for innovation. For example, the start-up ecosystem has grown in both number and global spread. As a result, established companies need to monitor more start-ups than before and therefore need to find new ways to identify, screen, and collaborate with start-ups.

    The significance and abundance of data are also increasing due to the growing digital data generated from social media, sensors, scholarly literature, patents, different forms of documents published online, databases, product manuals, etc. Various data sources can be used to generate ideas, yet, in addition to bias, the size of the available digital data is a major challenge when it comes to manual analysis.

    Hence, human-machine interaction is essential for generating valuable ideas where machine learning and data-driven techniques generate patterns from data and serve human sense-making. However, the use of machine learning and data-driven approaches to generate ideas is a relatively new area. Moreover, it is also possible to stimulate innovation using contest-driven idea generation and evaluation. However, the measurement of contest-driven idea generation processes needs to be supported to manage the process better. In addition, post-contest challenges hinder the development of viable ideas. A mixed-method research methodology is applied to address these challenges.

    The results and contributions of this thesis can be viewed as a toolbox of idea-generation techniques, including a list of data-driven and machine learning techniques with corresponding data sources and models to support idea generation. In addition, the results include two models, one method and one framework, to better support data-driven and contest-driven idea generation. The beneficiaries of these artefacts are practitioners in data and knowledge engineering, data mining project managers, and innovation agents. Innovation agents include incubators, contest organizers, consultants, innovation accelerators, and industries.

    Future projects could develop a technical platform to explore and exploit unstructured data using machine learning, visual analytics, network analysis, and bibliometric for supporting idea generation and evaluation activities. It is possible to adapt and integrate methods included in the proposed toolbox in developer platforms to serve as part of an embedded idea management system. Future research could also adapt the framework to barriers that constrain the development required to elicit post-contest digital service. In addition, since the proposed artefacts consist of process models augmented with AI techniques, human-centred AI is a promising area of research that can contribute to the artefacts' further development and promote creativity.

    Download full text (pdf)
    A toolbox for idea generation and evaluation
    Download (jpg)
    presentationsbild
  • 45.
    Ayele, Workneh Yilma
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Adapting CRISP-DM for Idea Mining: A Data Mining Process for Generating Ideas Using a Textual Dataset2020In: International Journal of Advanced Computer Sciences and Applications, ISSN 2158-107X, E-ISSN 2156-5570, Vol. 11, no 6, p. 20-32Article in journal (Refereed)
    Abstract [en]

    Data mining project managers can benefit from using standard data mining process models. The benefits of using standard process models for data mining, such as the de facto and the most popular, Cross-Industry-Standard-Process model for Data Mining (CRISP-DM) are reduced cost and time. Also, standard models facilitate knowledge transfer, reuse of best practices, and minimize knowledge requirements. On the other hand, to unlock the potential of ever-growing textual data such as publications, patents, social media data, and documents of various forms, digital innovation is increasingly needed. Furthermore, the introduction of cutting-edge machine learning tools and techniques enable the elicitation of ideas. The processing of unstructured textual data to generate new and useful ideas is referred to as idea mining. Existing literature about idea mining merely overlooks the utilization of standard data mining process models. Therefore, the purpose of this paper is to propose a reusable model to generate ideas, CRISP-DM, for Idea Mining (CRISP-IM). The design and development of the CRISP-IM are done following the design science approach. The CRISP-IM facilitates idea generation, through the use of Dynamic Topic Modeling (DTM), unsupervised machine learning, and subsequent statistical analysis on a dataset of scholarly articles. The adapted CRISP-IM can be used to guide the process of identifying trends using scholarly literature datasets or temporally organized patent or any other textual dataset of any domain to elicit ideas. The ex-post evaluation of the CRISP-IM is left for future study.

    Download full text (pdf)
    fulltext
  • 46.
    Ayele, Workneh Yilma
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Juell-Skielse, Gustaf
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    A Systematic Literature Review about Idea Mining: The Use of Machine-Driven Analytics to Generate Ideas2021In: Advances in Information and Communication: Proceedings of the 2021 Future of Information and Communication Conference (FICC), Volume 2 / [ed] Kohei Arai, Cham: Springer, 2021, p. 744-762Conference paper (Refereed)
    Abstract [en]

    Idea generation is the core activity of innovation. Digital data sources, which are sources of innovation, such as patents, publications, social media, websites, etc., are increasingly growing at unprecedented volume. Manual idea generation is time-consuming and is affected by the subjectivity of the individuals involved. Therefore, the use machine-driven data analytics techniques to analyze data to generate ideas and support idea generation by serving users is useful. The objective of this study is to study state-of the-art machine-driven analytics for idea generation and data sources, hence the result of this study will generally serve as a guideline for choosing techniques and data sources. A systematic literature review is conducted to identify relevant scholarly literature from IEEE, Scopus, Web of Science and Google Scholar. We selected a total of 71 articles and analyzed them thematically. The results of this study indicate that idea generation through machine-driven analytics applies text mining, information retrieval (IR), artificial intelligence (AI), deep learning, machine learning, statistical techniques, natural language processing (NLP), NLP-based morphological analysis, network analysis, and bibliometric to support idea generation. The results include a list of techniques and procedures in idea generation through machine-driven idea analytics. Additionally, characterization and heuristics used in idea generation are summarized. For the future, tools designed to generate ideas could be explored. 

    Download full text (pdf)
    fulltext
  • 47.
    Ayele, Workneh Yilma
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Juell-Skielse, Gustaf
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Social Media Analytics and Internet of Things: Survey2017In: Proceedings of the 1st International Conference on Internet of Things and Machine Learning, Association for Computing Machinery (ACM), 2017, article id 53Conference paper (Refereed)
    Abstract [en]

    Due to the emergence of social media, there is a paradigm shift in the area of information production, processing and consumption. Hence, investigation in the utilization of open social media data is a relevant research topic. The openness of data, social media data, enables innovation and societal value creation. Social media analytics is an evolving research domain with interdisciplinary methods that are common in data mining such as text mining, social network analysis, trend analysis, and sentiment analysis. Also, social media analytics deals with development and evaluation of frameworks and informatics tools to process noisy and unstructured social media data. On the other hand, Internet of Things (IoT) enables the utilization of digital artifacts with well-established solutions and allows things to be connected regardless of location and time. However, a literature review about social media analytics and IoT integration is missing. In this paper, we conducted a systematic literature review of social media analytics and IoT integration. The literature review indicates that there are fewer research works done in the area of social media analytics and IoT compared to Data Mining and IoT. This paper facilitates discussion and elicits research potentials in social media analytics and IoT integration.

  • 48.
    Azari, Amin
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Papapetrou, Panagiotis
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Denic, Stojan
    Peters, Gunnar
    Cellular Traffic Prediction and Classification: A Comparative Evaluation of LSTM and ARIMA2019In: Discovery Science: Proceedings / [ed] Petra Kralj Novak, Tomislav, Šmuc, Sašo Džeroski, Springer, 2019, p. 129-144Conference paper (Refereed)
    Abstract [en]

    Prediction of user traffic in cellular networks has attracted profound attention for improving the reliability and efficiency of network resource utilization. In this paper, we study the problem of cellular network traffic prediction and classification by employing standard machine learning and statistical learning time series prediction methods, including long short-term memory (LSTM) and autoregressive integrated moving average (ARIMA), respectively. We present an extensive experimental evaluation of the designed tools over a real network traffic dataset. Within this analysis, we explore the impact of different parameters on the effectiveness of the predictions. We further extend our analysis to the problem of network traffic classification and prediction of traffic bursts. The results, on the one hand, demonstrate the superior performance of LSTM over ARIMA in general, especially when the length of the training dataset is large enough and its granularity is fine enough. On the other hand, the results shed light onto the circumstances in which, ARIMA performs close to the optimal with lower complexity.

  • 49.
    Azari, Amin
    et al.
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Papapetrou, Panagiotis
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Denic, Stojan
    Petters, Gunnar
    User Traffic Prediction for Proactive Resource Management: Learning-Powered Approaches2020In: IEEE Global Communications Conference (GLOBECOM), IEEE, 2020, p. 1-6Conference paper (Refereed)
    Abstract [en]

    Traffic prediction plays a vital role in efficient planning and usage of network resources in wireless networks. While traffic prediction in wired networks is an established field, there is a lack of research on the analysis of traffic in cellular networks, especially in a content-blind manner at the user level. Here, we shed light into this problem by designing traffic prediction tools that employ either statistical, rule-based, or deep machine learning methods. First, we present an extensive experimental evaluation of the designed tools over a real traffic dataset. Within this analysis, the impact of different parameters, such as length of prediction, feature set used in analyses, and granularity of data, on accuracy of prediction are investigated. Second, regarding the coupling observed between behavior of traffic and its generating application, we extend our analysis to the blind classification of applications generating the traffic based on the statistics of traffic arrival/departure. The results demonstrate presence of a threshold number of previous observations, beyond which, deep machine learning can outperform linear statistical learning, and before which, statistical learning outperforms deep learning approaches. Further analysis of this threshold value represents a strong coupling between this threshold, the length of future prediction, and the feature set in use. Finally, through a case study, we present how the experienced delay could be decreased by traffic arrival prediction.

  • 50. Azari, Amin
    et al.
    Salehi, Fateme
    Papapetrou, Panagiotis
    Stockholm University, Faculty of Social Sciences, Department of Computer and Systems Sciences.
    Cavdar, Cicek
    Energy and Resource Efficiency by User Traffic Prediction and Classification in Cellular Networks2022In: IEEE Transactions on Green Communications and Networking (ITGCN), E-ISSN 2473-2400, Vol. 6, no 2, p. 1082-1095Article in journal (Refereed)
    Abstract [en]

    There is a lack of research on the analysis of peruser traffic in cellular networks, for deriving and following traffic-aware network management. In fact, the legacy design approach, in which resource provisioning and operation control are performed based on the cell-aggregated traffic scenarios, are not so energy- and cost-efficient and need to be substituted with user-centric predictive analysis of mobile network traffic and proactive network resource management. Here, we shed light on this problem by designing traffic prediction tools that utilize standard machine learning (ML) tools, including long shortterm memory (LSTM) and autoregressive integrated moving average (ARIMA) on top of per-user data. We present an expansive empirical evaluation of the designed solutions over a real network traffic dataset. Within this analysis, the impact of different parameters, such as the time granularity, the length of future predictions, and feature selection are investigated. As a potential application of these solutions, we present an ML-powered Discontinuous reception (DRX) scheme for energy saving. Towards this end, we leverage the derived ML models for dynamic DRX parameter adaptation to user traffic. The performance evaluation results demonstrate the superiority of LSTM over ARIMA in general, especially when the length of the training time series is high enough, and it is augmented by a wisely-selected set of features. Furthermore, the results show that adaptation of DRX parameters by online prediction of future traffic provides much more energy-saving at low latency cost in comparison with the legacy cell-wide DRX parameter adaptation.

1234567 1 - 50 of 544
CiteExportLink to result list
Permanent link
Cite
Citation style
  • apa
  • ieee
  • modern-language-association-8th-edition
  • vancouver
  • Other style
More styles
Language
  • de-DE
  • en-GB
  • en-US
  • fi-FI
  • nn-NO
  • nn-NB
  • sv-SE
  • Other locale
More languages
Output format
  • html
  • text
  • asciidoc
  • rtf